Zain Asif | AI Engineer & Senior Full Stack Developer

70%

LLAMA Model Accuracy

from 18%

65%

API Response Improved

80%

Faster Deployments

6+

Years Experience

Introduction

About Me

I'm a full-stack engineer and AI engineer with deep expertise in Python, AI/ML, and building scalable software solutions. I've led projects spanning E-commerce, Healthcare, and Retail, delivering complex systems like AI-powered Electronic Medical Records and Warehouse Management platforms.

At Turing, I worked on transformer-based model optimizations, RLHF pipelines, and agentic AI systems, reducing inference latency by 30% and aligning model outputs with ethical guidelines through adversarial red-teaming. Now focused on LLM evaluation and data engineering at Mercor.

Proficient in leveraging data-driven insights and innovative algorithms, I excel in crafting robust software architectures and efficient database solutions. Recognized for leading projects end-to-end and working effectively across diverse, cross-functional teams.

Location

Faisalabad, Pakistan

Experience

6+ Years

Availability

Remote / Global

What I Work With

Technical Skills

Languages

Python

Go

JavaScript

TypeScript

Ruby

PHP

Java

C/C++ SQL

Frameworks

Django

Flask

Node.js

React.js

Ruby on Rails

Laravel

Frontend & UI

HTML5

CSS / SCSS

Bootstrap

Tailwind

Figma

AI / Data Science

Prompt Engineering SFT DPO PPO RLHF Data Science Data Engineering NLP LLMs RAG Embedding Models Vector Databases Agentic AI Transformer Models Mixture of Experts

Databases

PostgreSQL

MySQL

SQL Server Database Design

Sequelize ActiveRecord

Amazon Web Services (AWS)

S3 EC2 RDS ECS Lambda CloudFront CloudWatch API Gateway SQS SNS IAM Route 53 Elastic Beanstalk ECR CodePipeline Secrets Manager

Cloud & DevOps

GCP

Azure

Docker

Heroku

Nginx

Linux Microservices CI/CD

APIs & Integrations

REST APIs Google API EasyPost API SP API / MWS Stripe QuickBooks API Square API NMI API Shopify Walmart API

Soft Skills & Leadership

Team Leadership Project Management Problem Solving Code Review Debugging Software Architecture Web Scraping Algorithms Client Communication Mentoring

Career Timeline

Work Experience

AI Engineer / Data Annotator

Mercor

Remote

Sep 2025 – Present

Analyzing LLM responses and taxonomy failure modes across diverse model outputs
Code debugging and SQL query optimization to improve model correctness
Using Claude Code and Copilot CLI to evaluate and test model behaviors
Applying SFT, RLHF and exhaustion testing to improve model accuracy and reliability
Video and image annotation for multimodal model training
Office documents (PDF, PPTX, DOCX) training data generation and evaluation
Setting up sandbox environments and DevOps for model training and data annotation

Python SFT RLHF LLMs SQL DevOps Claude Code

Data Engineer / Software Developer

Turing

Remote

Sep 2024 – Mar 2026

Architectural Optimization: Innovated transformer-based model enhancements (sparse attention, Mixture of Experts) to improve computational efficiency, reducing inference latency by 30%
Worked on data engineering of LLAMA model using Agentic and Agentless approaches, improving coding capabilities by 30% and increasing accuracy from 18% to 70%
Bias Mitigation & RLHF: Integrated RLHF and adversarial red-teaming to reduce toxicity by 40% in production models, aligning outputs with ethical guidelines
Built AI-first HR System (Turing Talent Bridge) for onboarding, off-boarding and transitions with integrations for Deel, Jibble, Docebo and OKTA. Deployed on GCP

Python TypeScript Go React Node.js PostgreSQL GCP DevOps

Data Annotation Engineer

SuperAnnotate

Remote

Dec 2024 – Jan 2026

Worked on AI data annotation and quality assurance for training large language models
Applied RLHF expertise to improve data labeling workflows and model alignment
Contributed to scalable data pipelines for machine learning training datasets

Python Data Engineering RLHF NLP

Full Stack Developer

Qbatch

Full-time

Jul 2021 – Oct 2024

Improved response time by 65% for existing APIs through optimization and refactoring
Built EMR (Electronic Medical Record) System with AI-based Test Recommendation System
Developed Everything Amazon WMS with AI Repricer, FBM and FBA support
Created microservices architecture and improved deployment time by 80%
Wrote algorithms to automate business processes and improved database structure
Integrated payment modules (Stripe, QuickBooks) and 3rd party APIs including Google API, EasyPost, SP API, and MWS
Improved usability through new UI designs and created comprehensive program flows

Python Ruby on Rails TypeScript Node.js React.js AWS Docker MySQL Stripe

Professional Freelancer

Fiverr

Remote

Jun 2018 – Aug 2021

Delivered projects ranging from small-scale scripts and scrapers to large-scale systems including Password Management Systems, LMS, and EMR
Handled full deployments and CI/CD pipelines across diverse technology stacks
Served clients globally across multiple industries and project sizes

Python Laravel Ruby on Rails Java C/C++ JavaScript PHP

Selected Work

Key Projects

sales.support

sales.support/features

Sales Support

sales.support · Nov 2021 – Jun 2024 · Team Lead

Featured Team Lead Qbatch

Full warehouse management platform with an AI Repricer that “increases sales by 30% or pays $100”. Includes Catalog Analyzer, All-in-one Vendors CRM, and Team Task Manager. Supports FBA, FBM, Shopify and Walmart with integrations across Google, QuickBooks, Stripe, EasyPost, SP API and more.

Ruby on Rails React Python Node.js GraphQL TypeScript AI Repricer

Key Impact

65% faster API response
80% faster deployments
DB & payment refactor

Turing

Meta Agentic / Agentless

Sep 2024 – May 2025

Data engineering on LLAMA to improve coding capabilities using Agentic and Agentless approaches. Boosted coding by 30% and increased accuracy from 18% to 70%. Large-scale team effort.

Python LLAMA Agentic AI Data Engineering

Turing

Turing Talent Bridge

May 2025 – Sep 2025

AI-first HR system for Turing handling onboarding, off-boarding and transitions. Integrates with internal projects plus Deel, Jibble, Docebo and OKTA. Deployed on GCP.

Python Go React GCP

Qbatch

Bluudo

Dec 2023 – May 2025

Rebuilt the project from single-item to multi-item ordering — effectively a full rewrite. Owned end-to-end full-stack development in TypeScript and React with PostgreSQL.

TypeScript React PostgreSQL Node.js

Qbatch

Amazon Catalogue Analyser

Nov 2021

Ruby script to fetch data from Amazon using SP APIs. Originally built on MWS (also authored by me), later migrated to the modern SP API for improved reliability and scalability.

Ruby SP API MWS Amazon

Freelance

Electronic Medical Record System

Jul 2020 – Jul 2021

Full-fledged EMR using Laravel that manages patient details with an integrated AI to recommend lab tests based on symptoms.

Laravel PHP AI/ML MySQL

Research

Traffic Congestion Research (GIS)

Mar 2020 – Jun 2020

Network Analysis on spatial dataset of King AbdulAziz University using ArcMap to help solve traffic congestion problems through GIS-based modeling.

Python ArcGIS GIS Data Science

My Toolbox

Tools & Platforms

Development

VS Code

Cursor

GitHub

Git

Docker

Postman

AI Platforms

Claude / Anthropic

OpenAI / ChatGPT

Google Gemini

Hugging Face

Productivity

Slack

Jira

ClickUp

Notion

Figma

Academic Background

Education

Bachelor of Science in Computer Science

National University of Computer & Emerging Sciences (FAST-NUCES)

Islamabad, Pakistan · CGPA 3.37 / 4.00 · 4 Years

Aug 2017 – Jul 2021

Academic Roles & Research

Teacher Assistant at FAST-NUCES — Sep 2018 to Aug 2019 (across multiple courses)
Spatial Analytics Research Group — Research on “Using GIS for Solving Traffic Congestion Problems at King AbdulAziz University” using ArcGIS (Mar 2020 – Present)

Relevant Coursework

Applied Machine Learning Artificial Intelligence Data Structures Design & Analysis of Algorithms Introduction to Cloud Computing Software Engineering Software Testing Object Oriented Analysis & Design Web Programming Discrete Structures Computer Programming Remote Sensing & GIS

Continuous Learning